Genomic imprinting is a gene regulatory process in mammals that distinguishes the parental origin of alleles at imprinted loci. DNA methylation is tightly linked to this process. Methylation imprints are parent-specific, are established in the gametes, and are stably maintained following fertilization. Correlations between methylation and expression differences in the adult suggest that DNA methylation is the inherited imprint which subsequently regulates parent-specific levels of expression. Disturbances in the imprinting process result in human congenital anomalies and cancer. The long-term objective of this research is to understand the molecular mechanism of genomic imprinting in mammals, thereby understanding mechanisms that control normal development and the transition to a cancerous state. This research will be pursued using the imprinted mouse RSVIgmyc transgene, which is a bona fide, portable imprinted gene whose methylation changes throughout gametogenesis and embryogenesis have been determined. The Specific Aims of this proposal are to characterize the cis-acting imprinting signal, and to analyze the functions of this signal in orchestrating the inheritance of allele- specific methylation patterns. Specific Aim 1. Characterization of the cis-acting requirement for genomic imprinting. The repetitive IgA switch recombination region of RSVIgmyc is the transgene's imprinting signal. Three experiments will be performed to test the hypothesis that the imprinting signal contains definable sequence characteristics. [l] Subsets of the RSVIgmyc imprinting signal, with slightly different sequence characteristics, will be tested for their imprinting ability. [2] Synthetic direct repeats of the RSVIgmyc imprinting signal will be tested for their ability to imprint the transgene. [3] Candidate imprinting signals from endogenous imprinted genes will be tested for their ability to imprint a version of RSVIgmyc in which the IgA imprinting signal has been removed. Specific Aim 2. Determination of the role of the imprinting signal in the inheritance of a transgene methylation pattern. To test the hypothesis that the imprinting signal is directly involved in the inheritance of gamete-derived methylation patterns, a non-imprinted version of RSVIgmyc in which the imprinting signal has been deleted will be studied. The methylation pattern of this RSVIgmyc modification will be examined during gametic and embryonic stages of the reproductive cycle, and the results compared to the developmental methylation changes of the intact RSVIgmyc transgene. Comparison of the two methylation profiles will lead to an understanding of the developmental and molecular roles of the imprinting signal. Specific Aim 3. Examination of the requirement of CpG dinucleotides within the imprinting signal. To address the requirement of CpG dinucleotides within the imprinting signal, all CpG dinucleotides within the signal will be mutated to TpG dinucleotides, and the modified construct tested for genomic imprinting. In addition, constructs designed and analyzed for genomic imprinting in Specific Aim l will be examined closely for a correlation between genomic imprinting and the presence of CpG dinucleotides in the imprinting signal.